Jack Dunn - Reclaiming Common Sense

This week was a slow week for economic news, if you were watching the news. First, and foremost, prayers go out to the people in Texas who are experiencing Hurricane Harvey. Hurricanes are nasty. This storm was supposed to fizzle out over the Yucatan Peninsula. Harvey had a different idea. Stay safe. This will have a major impact on millions of lives and on the country as a whole. We received information this week that was really quite good, other than Harvey. New home sales are improving. Existing home sales are improving. Both are doing better through the month of July than we have seen recorded since 2007. We are doing better than 2008, 2009, 2010, all the way through 2015 and 2016. So what was ignored this week?


(Aug. 23) The first column of the week was delayed due to the Solar Eclipse and due to research. These columns take research. It is more than regurgitating what is published by the government.  What is the raw data this month? What was the raw data last month? The first major economic report that was ignored this week was the New Home Sales Report. "July New Home Sales Shows Strength" details how even though there were fewer homes sold this July than during July 2016, we saw upward revisions to the prior months data, plural, and how we are on track for over 600,000 units sold, possibly 620,000, possibly higher. The authors of the report were misguided with their end of year estimate of 571,000.


(Aug. 24) Another major report released this past week was the weekly unemployment claims report. This was a major report because the unadjusted first-time claims number dropped back below 200,000 claims. We had not been under 200,000 claims for any week between June 1, 1974 and September 12, 2015.  We have added tens of millions of covered insured, potentially eligible workers, during that time. We tend to see our lowest level of first-time claims during the final week of September. "Crazy Weekly Unemployment Claims Data" details how if we were using the seasonal factors from the 1960s, 1970s, and early 2000s  that we could have seen a value of 226,000 seasonally adjusted first time claims reported. The only time that we have seen lower non-seasonally adjusted first-time claims this low during the third week of August was during 1967, 1968, and 1969. The article details a similar situation for the continuing claims data for the second week of August. This data is important because it means that we should see a drop in unemployment in the monthly Employment situation report that will be released next Friday. Ignore this weekly report at your own risk.


(Aug. 25) The July Existing Home Sales data was also essentially ignored this week. This column had published a July Real Estate Forecast column in advance of the release of the new construction data, the new home sales data, and the existing home sales data. It was thought, based on prior data that we could see stellar July existing home sales data. This July we sold the same number of homes as we did last July. This is good when you consider that we have an inventory level roughly 10% lower than last year (1.92 million versus 2.12 million.) It is more remarkable when you realize that this is the lowest July Inventory level since 1999. "July Existing Home Sales: Still Improving" details how we are on track for   5.6 million units sold - well above last year's 5.42 million units and comparable to the number of units sold during 2002. The best ever units sold years for existing homes were 2005, 2004, 2003, and 2002.


(Aug. 25)  Next week we will receive the "all-important"  August Employment Situation Report. This is an interesting month because the two data sets that are used to produce the report paint different pictures. The Current Population Survey (CPS) data from 2003 through 2016 reveal that we should trim part-time jobs, add full-time jobs and reduce the number of unemployed workers.  It also reveals that we should see both the non-seasonally adjusted (NSA) participation rate fall and the NSA U-3 unemployment rate fall. We normally see peak Summer participation during July. We could lose more part-time jobs than we gain in full-time jobs. The Current Employment Statistics (CES) data reveals that we normally add workers during  August. The article "August Jobs Report Forecast"  details how we had zero growth during August of 2015 and how that was manipulated to keep President Obama's "jobs" streak alive. This same article details also how we normally have a spike in workers during the August of the first year of a President's term in office, either a first term or a second term. We have seen spikes in workers of 0.15% during 1981, 1985, 1989, 1993, 1997, 2005, and 2013. The "devil is in the details." This could be rephrased to say "the devil is in the seasonal factors." Minor changes in the seasonal factor used to convert the NSA data to the SA data could mean 80,000 fewer workers being reported as being added to the economy. If the data from July is revised upward by 25,000 workers that effectively reduces the August gains by 25,000. Watch the growth rate, the seasonal factor and the revisions.


As Professor Slughorn say's the "Harry Potter and the Half-Blood Prince," "preparation is the prerequisite of all planning." In this case, planning is the forecast column. It is all about managing expectations. Much like Hurricane Harvey, if you know what to expect you can plan accordingly. Sometimes the economy takes a quick turn for the worse due to a "Black Swan Event." Hurricanes happen "every year." We have a Pacific Hurricane Season and an Atlantic Hurricane Season. We changed our annual vacation plans in North Carolina years ago after we had to wait to go to our rented house because of Hurricane Charlie. We now visit the Outer Banks during the early Summer instead of the late Summer. 


I recommend that anybody interested in the housing data or the jobs data actually dig into the data. Do not rely upon someone in Washington D.C. to produce a report without bias.  The data is improving.


"All of the data" is connected. If people are working full-time jobs then they may not need to work two jobs. If people decide to work two jobs then they may decide to choose to work a full-time job and a part-time job. We have been at or near record levels of people working two part-time jobs during any given month for the past few years.  If people lose one of their multiple jobs they do not receive unemployment benefits or even lesser employed benefits. If people are working fewer multiple jobs, or better jobs, then they may have more money to spend and more time to spend that money. More full-time jobs may lead to more people deciding to buy homes, new or existing. People who buy houses may need to buy new furniture, new electronics or appliances, or may undertake home improvement projects. Retail sales are improving. We are a consumption based economy. If sales improve then employers may decide to hire more workers.  More jobs, more income, more spending.


We have a failure to participate. There is a theory in economics called the "Phillips Curve." The thinking is that we have low inflation during times of high unemployment and high levels of inflation during periods of low unemployment. The problem is that when the concept was created we had a relatively stable participation rate.  We have seen the participation rate steadily fall since July of 2006. We have seen the Unemployment rate fall since July 2010.  If those people who are not participating were counted as unemployed, or effectively unemployed, the real unemployment rate would be in excess of 9.35%. We are seeing men reenter the workforce. We are seeing older people working longer.We are seeing that some sectors still have not recovered to pre-recession levels of workers. It takes days and sometimes weeks to dig into all of the data. This data is the "real news." Feel free to share this website with friends and co-workers.


It's the economy